{"id":145873,"date":"2023-03-27T19:00:56","date_gmt":"2023-03-27T17:00:56","guid":{"rendered":"https:\/\/www.hwcooling.net\/?p=145873\/"},"modified":"2023-06-28T13:12:46","modified_gmt":"2023-06-28T11:12:46","slug":"core-i5-13400f-complete-gaming-tests-b0-vs-c0-without-e-cores-too","status":"publish","type":"post","link":"https:\/\/www.hwcooling.net\/en\/core-i5-13400f-complete-gaming-tests-b0-vs-c0-without-e-cores-too\/","title":{"rendered":"Core i5-13400F complete gaming tests (B0 vs. C0). W\/o E cores too"},"content":{"rendered":"

The third and final Intel Core i5-13400F<\/a> CPU test with measurements beyond standard<\/a> will add to what the previous two have shown. We ran the two existing steppings \u2013 B0 (with Raptor Cove\/Raptor Lake P cores) and C0 (Golden Cove\/Alder Lake)<\/a> \u2013 through all the games we use to test processors, in four different display setups. In the case of stepping B0 even with modified CPU settings, with E cores disabled<\/a>.<\/p>\n

A brief look back (you may skip):<\/strong> From previous tests, you know that the Core i5-13400F processors’ characteristics vary a bit across steppings.Stepping B0 (S-Spec SRNBG) with the bigger die (with 8 and 16 E cores) is usually a bit faster in multi-threaded workloads<\/a>, but with higher power draw and it achieves lower efficiency<\/a>. However, this is no longer the case in single-threaded loads. Here, for the fairly higher power draw percentage, the stepping C0 from the Alder Lake generation<\/a>, which is built on the older Intel 7 manufacturing node (instead of Intel 7 Ultra \u2013 B0), achieves lower power efficiency. The most significant difference to the disadvantage of stepping C0 is then in idle operation.<\/a><\/div>\n

The results of this article follow up on game tests, of which we have only had a few. From testing ten games at 720p, 1080p, 1440p and 2160p, it’s possible to draw more objective conclusions. Whether it’s the difference in game performance due to different stepping (B0\/C0) or the (de)activation of the four “efficient” Gracemont cores<\/a>.<\/p>\n

\"\"<\/a><\/p>\n

“True” Raptor Lake is faster…<\/strong><\/h2>\n

On average, stepping B0 achieves 2\u20135% higher gaming performance than stepping C0<\/a>, but this comes at the price of slightly lower efficiency<\/a>. The power draw of the B0 variant can be higher possibly by up to 14%<\/a> (i.e. approx. 7 W).Of the games tested, the biggest speed difference in favour of stepping B0 is in CS:GO \u2013 up to 25%<\/a> (1080p@high), in DOOM Eternal \u2013 up to 11%<\/a>.<\/p>\n

There is also a bigger difference (up to 10%) in the minimum fps in F1 2020<\/a>, which have smaller drops with stepping B0<\/a>. In most games, such as Assassin\u2019s Creed: Valhalla<\/a>, Borderlands 3<\/a>, Cyberpunk 2077<\/a>, MS Flight Simulator<\/a> or Shadow of the Tomb Raider<\/a>, however, the difference is always less than 5% (and often only around 2%). But heads up, sometimes stepping C0 has the upper hand. Albeit always only symbolically, but we’ve nevertheless measured repeatedly higher fps in Assassin’s Creed: Valhalla<\/a> or Total War Saga: Troy<\/a> and from the perspective of minimal fps, in v Metro Exodus<\/a> as well.<\/p>\n

The difference in efficiency is at 2% in favour of stepping C0.<\/a> The latter is slightly slower on average across all games tested<\/a>, but also draws less power. And the lower power draw (compared to gaming performance) is the winner here<\/a>.<\/p>\n

\"\"<\/a><\/p>\n

… and little (E) cores don’t help much<\/strong><\/h2>\n

But perhaps the most interesting thing is that the Core i5-13400F’s highest gaming performance can be achieved with the E cores disabled<\/a> (i.e. with only 6 P cores). Of course, this is not always the case and cannot be generalized, but in any case the results show that at lower CPU load (at higher resolutions 1440p and 2160p<\/a>) the processor score is already a bit higher on average in the configuration without E cores<\/a>. They don’t reduce game performance though, and on the contrary, in games that can use part of the tasks assigned to E cores in addition to 6 P cores, running with them is faster. However, this occurs at lower resolutions<\/a> (with higher CPU load) and on average it’s up to 3%<\/a>.<\/p>\n

In F1 2020, where E cores have virtually no impact on fps, power draw is significantly higher with them, by 14%.<\/a> The percentage increase in power draw (+7%)<\/a> over game performance (+2%)<\/a> is also higher in Shadow of the Tomb Raider.But then again, in Total War Saga: Troy, power draw is 7% lower with E cores<\/a> at comparable speed (~201 fps)<\/a>. It’s individual, but mostly, on average, E cores in games mean lower efficiency<\/a>. This is also indicated by the gaming performance per unit of power draw ratio.<\/a> This is 4% higher after disabling the E cores, when the processor has only 6 P cores available.<\/a><\/p>\n

\"\"<\/a><\/p>\n

How did we get the results?<\/strong> All tests are carried out under strictly controlled conditions with very low tolerances in temperature or humidity differences. You can find out under what circumstances the tests take place on this page<\/a> and on the pages following it (chapter 3<\/a> to chapter 5<\/a>). Within the context of this article, however, it should be pointed out that power draw is measured with current clamps<\/a>. It is thus the clean value of power draw of the processor, undistorted by the other components.<\/div>\n

English translation and edit by Jozef Dud\u00e1\u0161<\/em><\/p>\n